University of Louisville ThinkIR: The University of Louisville's Institutional Repository

Electronic Theses and Dissertations

8-2006

Kinetics of sulfur and peroxide cured EPDM rubber aging in chloraminated water.

Jahnavi Valleru 1982- University of Louisville

Follow this and additional works at: https://ir.library.louisville.edu/etd

Recommended Citation Valleru, Jahnavi 1982-, "Kinetics of sulfur and peroxide cured EPDM rubber aging in chloraminated water." (2006). Electronic Theses and Dissertations. Paper 1481. https://doi.org/10.18297/etd/1481

This Master's Thesis is brought to you for free and open access by ThinkIR: The University of Louisville's Institutional Repository. It has been accepted for inclusion in Electronic Theses and Dissertations by an authorized administrator of ThinkIR: The University of Louisville's Institutional Repository. This title appears here courtesy of the author, who has retained all other copyrights. For more information, please contact [email protected]. KINETICS OF SULFUR AND PEROXIDE CURED EPDM RUBBER AGING IN

CHLORAMINATED WATER

By

lahnavi Valleru B. Tech, S.Y University College of Engineering, 2003

A thesis Submitted to the Faculty of the Graduate school of the University of Louisville in Partial Fulfillment of the Requirements for the Degree of

Master of Science

Department of Chemical Engineering University of Louisville Louisville, Kentucky

August 2006

KINETICS OF SULFUR AND PEROXIDE CURED EPDM RUBBER AGING IN

CHLORAMINATED WATER

By

Jahnavi Valleru B. Tech, S.V University College of Engineering, 2003

A Thesis Approved on

June 26, 2006

By the following Thesis Committee:

(Thesis Director)

ii DEDICATION

This thesis is dedicated to my parents

Mr. Muni Chandra Naidu Valleru

And

Mrs. Yasoda Valleru who have given me invaluable educational opportunities.

iii ACKNOWLEDGEMENTS

I would like to sincerely thank my academic director, Dr. Thomas L Starr, for his guidance throughout my graduate study and thesis work. I would also like to express my gratitude to Mr. Mark Schreck for his support during the research work and my practical training at Hexion Specialty Chemicals. I immensely appreciate the direction and help of

Dr. Thomas Rockaway and Dr. Gerold Willing for their invaluable suggestions and help with various aspects of research and results. I would also thank Ms. Rodica McCoy for helping me patiently in getting started with various instruments for my laboratory work.

Also, I would like to convey special thanks to Ranjith, for his understanding and patience during tough times; Rupa, for her endurance and a continuous support;

Rajmohan for his valuable inputs and suggestions to my thesis work.

IV ABSTRACT

KINETICS OF SULFUR AND PEROXIDE CURED EPDM RUBBER AGING IN

CHLORAMINATED WATER

J ahnavi Valleru

June 26, 2006

Elastomer degradation when exposed to chloraminated potable water is dependent on a number of poorly understood factors. Temperature and chloramine concentration are both believed to play roles, however, they have not been adequately defined. In order to estimate the material performance and predict the service life of elastomeric parts, knowledge of degradation modes, rate constants of the aging process and their activation energies is useful.

EPDM elastomers are used in numerous applications which require resistance to ozone attack and weather due to their stable, saturated polymer backbone structure. Accelerated aging experiments were conducted on both EPDM - Sand EPDM - P at three concentration levels and temperatures. Tensile properties are used to study the extent of degradation through a 30-day aging period which are used to estimate the crosslink densities of both the materials. A rate expression is developed based on the rate of change in cross link densities.

The rate constants are estimated and they are used to estimate the activation energies and frequency factors using an Arrhenius relationship.

v TABLE OF CONTENTS

PAGE

TITLE PAGE ...... i APPROVAL PAGE ...... ii DEDICATION ...... iii ACKNOWLEDGEMENTS ...... iv ABSTRACT ...... v LIST OF TABLES ...... viii LIST OF FIGURES ...... ix LIST OF CHARTS ...... x

CHAPTER

I. INTRODUCTION ...... 1

II. COMPOUNDING OF EPDM - SAND EPDM - P ...... 3

11.1 Nomenclature ...... 3

11.2 Polymer structure ...... 3

II.3 Polymer un-saturation ...... 4

II.4 Diene monomers in EPDM ...... 5

II.5 Manufacturing process ofEPDM ...... 7

II.6 Sulfur vulcanization ...... 9

II.7 Peroxide vulcanization ...... 12

III. EQUIPMENT AND PROCEDURES ...... 15

III. 1 Introduction ...... 15

III.2 Accelerated aging ...... 15

III.3 Testing conditions ...... 16

VI I1I.4 Sample materials and coupons ...... 17

IlI.5 Coupon exposure set-up ...... 19

I1I.6 Preparation of chloramine test solution ...... 22

III. 7 Elastomer tensile properties testing methodology ...... 25

IV. CONCEPTS OF RUBBER ELASTICITY ...... 27

IV.1 Kinetic theory of rubber elasticity ...... 27

IV.2 Activation energy ...... 29

V. PERFORMANCE RESULTS AND DISCUSSION ...... 31

V.1 Gaussian theory of elasticity ...... 31

V.2 Rate of change of cross link densities ...... 35

V.3 Model of rate kinetics ...... 39

V.4 Temperature dependence of rate and activation energies ...... 43

V.5 Prediction of service life ...... 46

V.6 Conclusions ...... 47

V.7 Recommendations ...... 48

VI. REFERENCES ...... 49

VII. APPENDIX 1...... 51

VIII. APPENDIX II...... 52

IX. CURRICULUM VITAE ...... 61

Vll LIST OF TABLES

TABLE PAGE

I1L1 Matrix of testing conditions ...... 16

I1L2 Description of tasks carried out and the conditions tested ...... 16

III.3 Composition of EPDM rubber material obtained from Ashtabula ...... 18

IlI.4 Procedure to prepare chloramine solutions of different concentrations ...... 22

V.1 Crosslink densities of EPDM-P, aged at 60ppm of chloramine concentration .... 33

V.2 Crosslink densities of EPDM-P, aged at 30ppm of chloramine concentration .... 33

V.3 Crosslink densities of EPDM-P, aged at 1ppm of chloramine concentration ...... 33

V.4 Crosslink densities of EPDM-S, aged at 60ppm of chloramine concentration ... .34

V.5 Crosslink densities of EPDM-S, aged at 30ppm of chloramine concentration .... 34

V.6 Crosslink densities of EPDM-S, aged at 1ppm of chloramine concentration ...... 34

V.6a Standard deviations of the respective CLDs for EPDM - P aged at 60ppm ...... 35

V.7 Rate of change of crosslink densities in 30 days for EPDM-P ...... 36

V.8 Rate of change of crosslink densities in 30 days for EPDM-S ...... 37

V.9 Extrapolated reaction rates compared with those of 60ppm for EPDM - P ...... 40

V.1O Rate constants of EPDM - P estimated by nonlinear regression ...... 41

V.11 Rate constants of EPDM - S estimated by regression ...... 42

V.12 Extrapolated reaction rates compared with those of 60ppm for EPDM - S ...... 42 2 V.13 Ea, A and R values of both the reaction steps for EPDM -P ...... 46

V.14 Ea, A and R2 values of the reaction steps for EPDM -S ...... 46

Vlll LIST OF FIGURES

FIGURE PAGE

III. 1 Dimensions of the rubber coupon used for tensile testing ...... 19

III. 2 Schematic diagram of the hot water circulation bath ...... 20

III. 3 Glass fixtures loaded with rectangles and tensile samples ...... 20

III. 4 Glass fixture with loaded samples in a stainless steel container with

chloramine test solution ...... 21

III. 5 Constant temperature circulating bath with digital temperature

display on the front...... 21

III. 6 Chlorine Titrimeter with electrode dipped into the test solution ...... 24

IV.1 A stress-strain curve showing that equation IV.1 is nonlinear (Hertz, 1991) ...... 29

IV.2 Extent ofreaction and the significance of activation energy ...... 30

V.1 Plot generated by Instron which matches the trend in figure IV.l...... 32

V.1a A plot of experimental data from Instron showing the linearity of the data ...... 36

IX LIST OF CHARTS

CHART PAGE

V.1 Rate of change of CLD of EPDM - P with respect to chloramine

concentration at three temperatures ...... 37

V.2 Rate of change of CLD of EPDM - S with respect to chloramine

concentration at three temperatures ...... 38

V.3 Percent change of hardness of EPDM - S during the 30-day aging

period at 1ppm chloramine concentration and three temperatures ...... 39

V.4 Arrhenius plot of low concentration reaction rate constants vs. lIT

for EPDM - P ...... 44

V.5 Arrhenius plot of low concentration reaction rate constants vs. liT

for EPDM - S ...... 45

x CHAPTER I

INTRODUCTION

All structural materials display a degree of elasticity, but elastomers as a materials class are distinct. They can be loosely defined as macromolecular in structure, stretchable under low stress to at least twice their original length, and having a high degree of shape memory. They have become indispensable to the manufacturing industry and in particular industry producing valves, pipes, fittings, and the host of appurtenances required by the water works field.

Since the early 1900s, water systems throughout the US have included disinfection in their purification process. Originally, free chlorine was used as the only disinfectant, however, due to regulatory, economic and health issues, many systems have switched to chloramines in the distribution system. The chloramine residual is generally preferred because of its persistence in the distribution system and its tendency to form fewer disinfection by-products. In some situations, however, chloramine residuals have had significant negative impacts on the elastomeric parts of water distribution infrastructure.

Elastomer degradation when exposed to potable water is dependent on a number of poorly understood factors. Simmons and Evanson (1988) conducted research that dealt directly with potential mechanisms of chloramine attack. They were able to demonstrate that the chloramine effects on the raw polymer were minimal and that the vulcanizate impact was substantial, including loss of tensile strength and extensive swelling. This was true for Nitrile, Butyl and EPDM based elastomers. They were also able to demonstrate that the extent of attack was dependant on the form of the cross-linking; the more common sulfur based vulcanizates were more susceptible to attack than elastomers cured in a peroxide process.

To further investigate the impact of the vulcanization process on chloramine resistance, sulfur cured and peroxide cured EPDM materials are selected for failure analysis and degradation rate assessment. A background of EPDM rubber compounding and molecular structure is discussed in chapter 'Compounding of EPDM - Sand EPDM

- P'. These elastomers are subjected to accelerated aging in chloramine solutions at three concentration levels and three temperatures. The design and set-up of the accelerated tests are presented in the chapter 'Equipment and procedures'. Also, the operating procedure of Instron tensile testing machine which is used to measure the tensile properties is explained in this section.

Deterioration in tensile properties is measured at set intervals during a 3D-day aging period. Tensile properties are used to determine the cross-link densities of both the materials. The rates of change are further analyzed to model degradation reaction kinetics and estimate the respective activation energies. In addition to a comparison of activation energies and reaction rates of EPDM - Sand EPDM -P, the mode of failure is also discussed in the 'Performance results and discussion' chapter.

2 CHAPTER II

COMPOUNDING OF EPDM - SAND EPDM - P RUBBER

ILl Nomenclature

Ethylene - propylene rubbers are random copolymers of the two hydrocarbons ethylene and propylene with the ethylene varying from 40% to 70% by weight. Small

quantities of a third monomer are added to produce ethylene-propylene terpolymers which can be vulcanized in the usual way. EPDM follows a nomenclature convention endorsed by the American Society for Testing Materials and the International Standards

Organization and applies to the more common, sulfur vulcanizable product which

includes in the rubber molecule a minor percentage of a diene monomer in addition to ethylene and propylene. The basis for the letter designation, EPDM, is: Ethylene,

Propylene, Diene and Methylene respectively. In this case, the methylene molecules are the repeating units (CHz), or "Vertebrae", in the "Spine" of the polymer.

11.2 Polymer Structure

The structure of the regular, alternating amorphous copolymer of ethylene

HzC = CHz and propylene

can be written as,

3 Ethylene Propylene

This structure is remarkably similar to the structure of natural rubber, cis 1, 4 polyisoprene:

CH3 H2 H I H~ C-C==C-C -f n

It is not surprising; therefore, that the regular, alternating copolymer of ethylene and propylene is a decidedly rubbery material since its classical structure so closely approaches that of the first useful elastomer, natural rubber.

The equimolar structure of ethylene and propylene, as shown previously, is probably not achieved in the present commercial ethylene/propylene rubbers. The compositions of the commercial materials are generally given on a weight percentage basis. Although the rubbery properties of the ethylene/propylene copolymers are exhibited over a broad range of composition, the commercial products are generally in the weight percentage range of 50/50 or 75/25 ethylene/propylene.

II. 3 Polymer Unsaturation (EPDM):

The structure of EPDM as illustrated previously shows this to be a saturated synthetic rubber. There are no double bonds in the polymer chain as there are in the case of natural rubber and in most common commercial synthetic rubbers (e.g., SBR, CPBR,

NBR, etc.,). The main chain unsaturation in these latter materials introduces points of

4 weakness. When exposed to the degrading influences of light, heat, oxygen and ozone, the unsaturated rubbers tend to degrade through mechanisms of chain scission and crosslinking involving the points of carbon-carbon unsaturation. Since EPM does not contain any carbon-carbon unsaturation, it is inherently resistant to degradation by heat, light, oxygen and, in particular, ozone.

The double bonds in natural rubber and the common polydiene synthetics, aside from rendering these elastomers susceptible to environmental degradation, are essential to their curing into useful rubber products using conventional chemical accelerators and sulfur. The entire world's rubber industry is based on the vulcanization chemistry which was first demonstrated by Charles Goodyear in 1839 and involves carbon-carbon double bonds in a sulfur crosslinking reaction. EPM, a saturated elastomer, cannot be cured or crosslinked using the long-established manufacturing practices and chemicals pertinent to the unsaturated rubbers. A more commercially attractive product would be one which retained the outstanding performance features (e.g., heat, oxygen, ozone resistance) and which included some carbon-carbon unsaturation from a small amount of an appropriate diene monomer to accommodate it to conventional sulfur vulcanization chemistry.

II. 4 Diene monomers in EPDM:

Efforts to introduce one common diene monomers into the EPDM molecule have been unsuccessful. Searches for appropriate nonconjugated dienes resulted in the discovery of fifty such chemicals. The lowest molecular weight straight chain diolefin which meets the requirements is 1, 2 hexadiene:

H3 H H H2 H H2 C-C=:C-C-C==C

5 When this chemical is introduced with ethylene and propylene, the terminal double bond is active with respect to polymerization. The internal unsaturation is passive at this stage but remains in the resulting terpolymer as a substituent, or pendant, location for sulfur vulcanization: --I--Hz Hz lH, Hz H H, ..2\- \ C-C===c-C-j-C In

CH2 I CH II CH I CH1

Ethylene propylene 1, 4 hexadiene

The resulting Ethylene/propylene/I, 4 hexadiene terpolymer (EPDM) is an important commercial material. Dicyclopentadiene is used in certain grades of EPDM which are made by a number of manufacturers. It enters the polymer readily with much higher polymerization efficiency than 1, 4 hexadiene. However, the most widely used diene in current commercial EPDM is ethylidene norbornene (ENB). H H3 c C W -

As with the other bridged ring dienes, ENB shows a high rate of polymerization through the double bond in the bridged ring. The substituent internal double bond is also very active with respect to sulfur crosslinking. The structure of EPDM containing ENB

IS, H2 H2 1H3 H2v- -C-C -C-c ( n

CH I CH3

6 The most common commercially used dienes in EPDM are 1, 4 hexadiene, dicyclopentadiene (DCPD) and ethylidene norbornene (ENB). These enter the polymer through the double bond which is more active in this function and leave the second double bond substituent or external to the polymer chain where it is then available to function in the classical chemistry of the sulfur vulcanization of rubber.

II. 5 Manufacturing Process of EPDM

11.5.1 Feed Specifications:

Ethylene is copolymerized by saturating the liquid reaction mixture with a given concentration of propylene and maintaining the saturation at this level throughout the reaction. The composition of the copolymer being formed can be maintained by monitoring the composition of the off-gas from the reaction and adjusting the ratios of the ethylene and propylene in the inlet streams.

When incorporating a diene such as dicyclopentadiene to give a curable polymer, all of the dicyclopentadiene may be present before the catalyst is added or alternatively, part or all of the dicyclopentadiene may be added during the copolymerization process.

This introduction may be continuous or periodic during the reaction.

II.S.2 Temperature and Pressure:

The copolymerization reaction may be carried out over a wide range of temperatures. In general, as the temperature of the reaction increases, the catalyst life and the molecular weight of the copolymer decrease. Temperatures within the range of O°C

25°C to about 80°C may be used. Pressures from 14 to 1,000 psig can be used. In the case of ethylene - propylene mixtures, the preferred polymerization pressures range from 50 to 200 psig.

7 II.5.3 Reaction Medium:

The copolymerization process is preferably carried out in an inert liquid organic diluent which is a solvent for the polymerization system. To obtain a co-polymer product of homogenous composition throughout, the diluent should be one that is a solvent not only for the monomers being copolymerized but also for the copolymer that is produced.

In addition, it should also be a solvent for the catalyst so that the entire copolymerization reaction mixture is homogenous throughout the copolymerization process.

Suitable diluents for the copolymerization are, in general, the hydrocarbon solvents, i.e. aromatic, alicyclic and aliphatic hydrocarbons and chlorinated aromatic, alicyclic and aliphatic hydrocarbons and mixtures thereof. Examples of such diluents that may be used include hexane, heptane, octane, nonane, decane, benzene, toluene, methylene chloride, carbon tetrachloride tetrachloroethylene, chlorobenzene, dichlorobenzene etc.

11.5.4 Catalyst:

One of the criteria in carrying out the copolymerization process and producing a homogenous product of uniform composition and narrow molecular weight distribution is the catalyst that is used for the copolymerization reaction. It must, as a rule, be completely soluble in the reaction mixture.

A vanadium oxytrichloride, aluminum alkyl, iodine mixture is a useful catalyst. The catalyst systems are made by contacting vanadium tris (acetylacetonate) with selected organoaluminum compounds in the presence of selected liquid halogenated aliphatic hydrocarbons which serve as the reaction media when these catalyst systems are used to polymerize the monomers.

8 11.5.5 Product recovery:

The polymerization may be stopped by deactivating the catalyst with an alcohol such as isopropanol. The polymers formed are isolated by conventional filtration after the slurry has been processed. In general, the polymer is treated with an aqueous mineral acid to remove vanadium and aluminum salts; the solution or slurry is thereafter washed with distilled water until it is acid - free.

Improved catalyst removal may be attained by adding to the solution, as it recovered from the reactor, a surfactant, a chelating agent, and water, and then agitating the mixture in order to emulsify the solution. Once the emulsion has been formed, the gel will rise to the surface in a form resembling oatmeal, and may be readily removed by skimming or centrifuging. After removal of the gel, the emulsion is broken by heating to a temperature in excess of 65°C, preferably about 88°C, and the aqueous phase, containing dissolved catalyst residues, is separated from the hydrocarbon phase.

11.6 Sulfur Vulcanization

After the EPDM rubber compound has been properly mixed and shaped into blanks for molding, or calendered, extruded, or fabricated into a composite item, such as a tire, it must be vulcanized by one of many processes. During vulcanization, the following changes occur:

1. The long chains of the rubber molecules become crosslinked by reactions with the

vulcanization agent to form three- dimensional structures. This reaction transforms

the soft plastic - like material into a strong elastic product.

9 2. The rubber loses its tackiness and becomes insoluble in solvents and is more

resistant to deterioration normally caused by heat, light and aging processes.

11.6.1 Mechanism:

Since EPDM rubber contains unsaturation, vulcanization with sulfur is possible, and it is in general the most common vulcanizing agent used. With sulfur, crosslinks and cyclic structures of the following type are formed:

Generally, x in an efficient accelerated curing system is about 1 or 2, with little or no cyclic groups formed. In inefficient systems x equals up to 8 and many cyclic structures are formed. The total amount of sulfur combined in these networks is usually called the "Coefficient of Vulcanization" and is defined as the parts of sulfur combined per one hundred parts of rubber by weight. For most rubbers, one crosslink for about each

200 monomer units in the chain is sufficient to produce a suitable vulcanized product. It is these amounts of cyclic sulfur (y) and the excessive sulfur in the crosslinks (x) which contribute to the poor aging properties of the vulcanizates.

Three agents essential to the sulfur vulcanization on EPDM are sulfur, a metal oxide, and a vulcanization accelerator. Of the metal oxides examined, only zinc and cadmium oxides are of practical significance. Zinc oxide is the preferred activator because it is more efficient, lower in cost, and less hazardous to use.

The best accelerators for the vulcanization of EPDM are also commonly used in natural rubber. However, their specific behavior differs from their response in rubber.

10 The relatively slower rate of vulcanization of EPDM requires a higher accelerator concentration, higher curing temperatures, or both. The most active accelerators include

2-mercaptobenzothiazole, thiuram sulfides, dithiocarbamates, and their simple derivatives. Of the three classes, the thiuram sulfides and dithiocarbamates are generally preferred because they produce rapid curing without scorching and do not over cure on long cure cycles. For many uses, however, 2-mercaptobenzothiazole and its derivatives, alone or in combination with thiurams or dithiocarbamates, provide adequate acceleration with processing safety.

EPDM polymers are amorphous and are similar to other non-crystallizing elastomers in the low tensile strength of their gum vulcanizates. However, high tensile strength is easily obtained by the incorporation of reinforcing fillers. The fillers commonly used in the rubber industry are carbon blacks, clay, calcium carbonates, calcium silicate, fine particle tales and silicas. Carbon blacks are preferred for maximum reinforcement.

Vulcanization is normally accomplished by applying heat for a specified time at the desired level. The most common methods for vulcanization are carried out in their molds held closed by hydraulic presses and heated by contact with steam-heated platens, which are a part of the press in open steam in an autoclave, under water maintained at a pressure higher than that of saturated steam at the desired temperature, in air chambers in which hot air is circulated over the product, or by various combinations of these methods. The vast majority of products are sulfur-cured; that is, sulfur crosslinks join the rubber chains together as described above.

11 The time and temperature required for vulcanization of a particular product may be varied over a wide range by proper selection of the vulcanizing system. The usual practice is to use as fast a system as can be tolerated by the processing steps through which the material will pass without "scorching", that is, without premature vulcanization caused by heat during these processing steps. Rapid vulcanization affects economies by producing the largest volume of goods possible from the available equipment. This is particularly the case for products made in molds, because molds are costly, and their output is determined by the number of heats which can be made per day.

The rate of vulcanization increases exponentially with an increase in temperature; hence the tendency is to vulcanize at the highest temperature possible. In practice, this is limited by many factors, and the practical curing temperature range is 260 - 340°F (127 -

171°C). There are numerous exceptions both below and above this range, but it probably covers 95% of the products made.

Finishing operations following vulcanization include removal of mold flash, sometimes cutting or punching size, cleaning, inspection for defects, addition of fittings such as valves or couplings, painting or varnishing, and packing.

11.7 Peroxide Vulcanization

Peroxide vulcanization of rubber dates back to about 1914. Rubber was cured with benzoyl peroxide. However, except for limited use with silicones, this peroxide has not found a place in rubber vulcanization. Then, about 1950, di-tertiary butyl peroxide was· found to give much better quality rubber vulcanizates than benzoyl peroxides. This peroxide was so volatile; however, that extreme care was required even in laboratory compounding to avoid excessive loss of peroxide during curing. Shortly thereafter,

12 dicumyl peroxide (Hercules Di- cup®) was found to have a good combination of physical and chemical properties for the compounding and curing of rubber.

Other peroxides are now available so that the rubber compounder may choose peroxide which gives a faster cure or one that may be used in compounds processed at higher temperatures. Among the peroxides available, however, an improvement in either of these two properties may be obtained only by a corresponding sacrifice of the other.

That is, processability at higher temperatures is obtainable only with peroxides which are slower curing at a given temperature. Although the decomposition products from the cross-linking agent depend on the peroxide used, the mechanism as related to the elastomer is essentially the same.

11.7.1 Mechanism:

In the absence of other additives, peroxide vulcanization consists almost exclusively of creating carbon - to - carbon bonds between the polymer chains. That is, the peroxide itself does not become a part of the cross-link in the polymer. The following reaction shows how a typical peroxide used in rubber vulcanization decomposes to form free radicals which cause cross-linking.

C C C I I ~-C-O:O-C-~ t, • 2 ( 0 ) \==./ I I \==./ ~ )-7- C C C

Homolytic Cleavage of Peroxide I

First the dicumyl peroxide undergoes a homolytic cleavage producing cumyloxy radicals.

Some of the cumyloxy radicals then abstract active unsaturated hydrogens from the polymer, producing polymer radicals which ultimately couple through carbon-carbon

13 bonds to produce a cross-linked form of the polymer. The remaining cumyloxy radicals decompose to acetophenone and methyl radicals. These methyl radicals also can abstract hydrogens from the polymer.

H c I I __ C-C-C-C-- ...... ~-C-OH+ I "=J I c C

-- C-C-C-C--• I Polymer Radical Formation C

The principal reaction products of dicumyl peroxide are cumyl alcohol, acetophenone, and methane. The coupling reaction which occurs during peroxide vulcanization of polymers is shown below:

C C I I -- C - C - C - C - C - C -- -- C - C - C - C - C - C -- • ---..... I -- C - C• - C - C - C - C -- __ C-C-C-C - C -C-­ I I C C

Although many other radicals probably can and do form, the important reaction of the radicals is coupling, as shown. The carbon - to - carbon coupling mechanism shown probably accounts for the increased oxidation stability of peroxide - cured rubber compared with sulfur - cured unsaturated polymers. Sulfur crosslinks are reported to be more sensitive to oxygen than are carbon - to - carbon bonds even in the age-resistant rubber products resulting from vulcanization with thiuram accelerators.

14 CHAPTER III

EQUIPMENT AND PROCEDURES

III. 1 Introduction

The unique properties of elastomers require special adaptations of the standard mechanical properties tests. The extent of elongation at failure, which directly relates to desirable elastomeric characteristics, is frequently a very useful performance parameter.

The extent of elongation also gives an indication of the quality and degree of polymeric cross-linking (vulcanization). All the testing methodologies used for the project are adopted from the standard procedures of American Society for Testing and Materials

(ASTM).

III. 2 Accelerated Testing

The time rate of chloramine-induced degradation of elastomers is controlled by both temperature and concentration levels. The acceleration study systematically varied these components and performed surface roughness, tensile strength, elongation and hardness tests to quantify the degradation. It was understood from the past research that exposure to nominal chlorine/chloramine concentrations of distribution systems would not produce meaningful differences in the performance characteristics of the elastomeric test materials. Therefore, an accelerated testing program using high-strength chlorinelchloramine solutions at elevated temperatures was devised.

15 III. 3 Testing Conditions

The ASTM procedure for accelerated elastomer degradation testing specifies a standard 70°C temperature and a 50ppm concentration for testing. While the ASTM protocols were followed for chloramines testing, the testing program was further explored by supplementing the ASTM test procedures to assess temperature and concentration effects. The standard ASTM test procedures were supplemented with additional accelerated degradation tests performed at temperature and chloramines concentration as defined in the following matrix.

x X X

X X ASTM

X X X

Temperature Table 111.1 Matrix of testing conditions

Nine different exposure conditions served as the basis of the accelerated life-cycle testing program. These nine conditions were conveniently divided into 3 test programs,

A, Band C as follows:

Chloramine Task Temperatures Concentrations

A 1 ppm 22°C, 45°C and 70°C

B 50 ppm 22°C, 45°C and 70°C

C 100 ppm 22°C, 45°C and 70°C

Table 111.2 Description of tasks carried out and the conditions tested

16 Test solution conditions are checked daily by measuring the pH and chloramine

concentration, and because chloramine is constantly consumed via reactions with the

elastomeric materials, the solutions were drained and replaced every 24 hours. Each test

program ran for 30 days. At the end of 3, 6, 12, 20 and 30 days the test coupons were removed from the test solutions and quantitatively assessed for degradation with the help

of physical and chemical tests.

III. 4 Sample materials and test coupons

Molded rubber samples are obtained from Ashtabula Rubber Company for the

purpose of the research. An agreement with Ashtabula Rubber Company was reached to

supply EPDM elastomer formulations (composition presented in table 111.3), based on a preliminary assessment of the elastomer variety in use and currently available to water distribution systems that will be used to represent "un-aged" conditions. These generally represent "R-Class" materials as defined by ASTM D2000. Elastomers from higher classifications (class M, Q and U) generally include specialty or propriety polymers and thus mayor may not be representative of base materials. These materials where received

in the form rectangular rubber slabs of size 6" x 7" x 0.80". These slabs are then cut into

coupons according to the test protocol.

Test coupons were divided into two types based on the type of tests conducted on them. For tensile testing, the specimens were cut into a dumbbell shape and for hardness,

change in mass and volume tests, they were cut into rectangles. These rectangles are removed on the test days for measurement and are replaced at the end of the testing.

These are called the reusable samples. The tensile samples are strained to failure on the test day to measure the tensile properties.

17 Ingredient PHR Function

Nordel1070 100 DuPont -1 ,2polybutadiene

FEFN550 100 Carbon black

Sun 22800 (or equiv.) 110 Extending oil

Sulphur 2 Primary vulcanizing agent

ZNO 5 Curing system additive (activator)

Stearic Acid 2 Curing system additive

Altax/Butyl ZlM Tuads 4.8 Accelerants

Table II!. 3 Composition of EPDM rubber material obtainedfrom Ashtabula

The reusable samples were cut in 2" x I" x .080" thick per ASTM D-3182. Tensile specimens were cut using a standard dumbbell specimen type C Die per ASTM D412.

The specimen die cutting surface was carefully maintained because the cutting surface is required to ensure a smooth edge on the cut specimen. The shape of the tensile specimen avoided uneven breakage and the tendency to slide out of the testing grips. The narrow neck of the dumbbell specimen ensured a consistent breakage pattern, and the coupon's

splayed ends gave substantially more grip area relative to the neck cross-section.

Following ASTM procedures, the coupons were cut lengthwise along the grain (the direction of extrusion). Shown in figure IlL 1 is the typical dumbbell coupon dimensions:

18 ;'--'-'-/2-"~ -·-1 :- ~ :;~: -1------;°1 ,...----... + + T 1"

1/4" 1/2" ~~

Figure II!.i Dimensions of the rubber coupon usedfor tensile testing

III. 5 Coupon Exposure Setup

Glass fixtures with spacers were used to suspend the coupons in solutions. The coupons were loaded onto the glass fixture and were placed in the test solution at the test temperature in a constant temperature circulating bath. This arrangement gave maximum exposure of rubber coupons to solution. Also, this design addressed a couple of serious concerns: first, the maintenance of equal fluid motion across the elastomer coupons and second, provision for ready inspection and removal of the coupons throughout the exposure sequence. A schematic representation of the setup is shown in figure IIl.2.

Each rubber type is placed in a stainless steel container with the chloramine test solution and the container is placed in the constant temperature circulating bath to maintain a constant temperature. A digital display on the front of the bath shows the temperature of the medium inside (figure IlLS).

19 "---- SS Cover

6 SS "Micro" Environments

NH2CI Solutions Poiymers Segregated

4 9jj.5" x 5 314"

Samples on glass rods Hot H20 Circulating C'ap approx 24

Figure III. 2 Schematic diagram of the hot water circulation bath

Figure III. 3 Glass fixtures loaded with rectangles and tensile samples

20 Figure Ill. 4 Glass fixture with loaded samples in a stainless steel container with

chloramine test solution

Figure III. 5 Constant temperature circulating bath with digital temperature display on

the front

21 III. 6 Preparation of the chloramine test solution

All the test solutions were prepared from the same de-ionized (DI) water having an initial resistivity of 18.3 Mega Ohms. Chlorine was added to the DI water in the form of sodium hypochlorite (NaOCI); ammonia was added as ammonium chloride. To provide uniform excess, chlorine and ammonium were added to the DI water in the ratio of 4 to 1 weight ratio of chlorine to ammonium. The pH adjustment was achieved by adding 7.0pH phosphate buffer to the chloramine solutions. The detailed procedure to prepare a

100ppm chloramine solution is given below:

1. 9.0 L of DI water is taken in a container for each of the three temperatures, 22C,

45C and 70C.

2. 36 mL of 7.0 pH phosphate buffer is added to the water.

3. NaOCI is added per the table below to the solution.

4. NH40H is added to the solution per the table below.

5. 18 - 20 mL of 7.0 pH buffer is added to the solution to bring down the pH to 8.3.

Avg. Initial 7.0 pH NaOCI/ N~OHl 7.0 pH Temp °C Cone. Cone. Buffer/ L L L Buffer/ L 70 60 ppm 120 ppm 4ml 2.2ml 3.8 ml 2- 8 ml

45 60ppm 90ppm 4ml 1.7ml 2.7 ml 2- 8 ml

22 60ppm 60ppm 4ml 1.1 ml 1.8 ml 2- 8 ml

Table IlI.4 Procedure to prepare chloramine solutions of different concentrations

III. 6.1 Checking the concentration of chloramine solution:

Chloramine concentration is determined by an amperometric titration utilizing phenyl arsine oxide as the titrant. When the titrator cell is immersed in a sample

22 containing chlorine, current is generated. As phenylarsine oxide is added, the chlorine present as hypochlorite is reduced and the generation of current decreases proportionately due to the diminishing of the reducible species. When chlorine is present as chloramines, potassium iodide is added, releasing iodine which is titrated in a similar manner. The iodine content is calculated in terms of free chlorine. The instrument used is a Fisher CL

Titrimeter Model 397. A detailed procedure is given below:

1. The electrode is rinsed prior to the titration using a wash bottle with DI water.

2. For a 100 ppm nominal solution, 2 ml of sample is added to 98 ml of DI water

in the beaker provided with the instrument, for a total sample size of 100 ml.

(used 25: 1 dilution for 50 ppm solutions and no dilution for 1 ppm solutions)

3. To the diluted solution above, 1 ml of potassium iodide and 1 ml of 4.0 pH

buffer are added, always in this order.

4. The electrode is lowered down into the solution and the instrument should be

set for "Total" chlorine residual.

5. Using a pipette, phenylarsine oxide solution, the titrant is dispensed slowly

and during the delivery the current in the watch meter decreases.

6. The end point will be noted when the addition of titrant causes no further

change in the current. The pipette reading prior to the last delivery is used in

the equation below to determine total chlorine residual in ppm.

7. Equation: Chlorine residual (ppm) = 200 D (AlV), where D = the dilution of

the solution for the titration sample, A = ml of phenyl arsine oxide solution

required, and V = sample used in ml.

23 Figure III. 6 Chlorine Titrimeter with electrode dipped into the test solution

Freshly prepared chloramine solutions are then distributed into the stainless steel containers containing the glass fixtures loaded with rubber samples in each. These containers are then kept in the circulating baths to maintain temperature. The solutions are freshly prepared and replaced every 24 hours because chlorine is constantly consumed via reactions with the elastomeric materials and concentration drops considerably. These reactions on the elastomers consumed a substantial amount of the available chlorine (both free and combined) in the exposure solutions and usually altered solution pH as well. Significant variation in the chlorine exposure level and pH were unavoidable. To maintain the desired exposure conditions, it was necessary to monitor the baths daily, changing the solutions.

The chloramine levels were relatively stable and easier to maintain at Ippm than at

50ppm or lOOppm. The daily fluctuations of lOOppm were as high as 70ppm and for

50ppm they were 20ppm. This rapid chloramine consumption foreshadowed the

24 reactivity of these species relative to elastomer degradation. The rapid depletion of the chloramines was indicative of the impact they exerted on the elastomeric materials.

III.7 Elastomer tensile properties testing methodology

The attack on elastomeric structures can proceed by at least two mechanisms.

Depending on materials and vulcanization methods, the attack may concentrate on the polymeric cross-links formed during the vulcanization process, or, if the oxidant is powerful, it may actually rupture the polymeric backbone (in addition to the cross-links).

In either case, the weakened structure manifests itself in a variety of ways: The surface may break across lines of tension, forming cracks that are visible to the naked eye; the tensile properties of the material may be significantly degraded; or the structure may lose its resilience and become embrittled. The loss of polymeric structural integrity may also contribute to water absorption and a physical swelling of the material. Such changes may alter shape and can be accompanied by a loss of the elastomeric fillers (released due to the breakdown of the polymer encapsulation).

Because elastomers are most commonly utilized for their flexure and shock dampening properties, the elastomeric tensile property of greatest concern to manufacturers is usually elongation (strain). The ultimate elongation at failure is typically a more useful tensile property than failure stress (ultimate load). Stress and strain are closely related properties and are both determined by the intrinsic qualities of the polymeric structure. Oxidant attack weakens this structure, bringing about profound changes in the elastomer's load-bearing capacity - the degradation of which is often as great as or greater than the impact on ultimate elongation.

25 Tensile strength was determined per ASTM D 412C. Coupons were tested to failure on a model 4505 Instron Universal Testing system with a 2603-80 balanced, long-travel elastomeric extensometer installed. The load cell is 1.0 kN - 224.8 lbs. The standard coupon elongation rate was set at 20 in. Imin. Stress was determined based on the unloaded cross-sectional area of the coupon neck, and strain was calculated as percent of elongation along the neck of the coupon.

A brief procedure of tensile testing cycle is given below:

1. The dumbbell test coupons are removed from the test solution and are lightly

blotted with lint free paper to remove moisture.

2. The Instron machine's mainframe power is turned on and the tensile sample is

carefully inserted into the pneumatic clamps.

3. Extensometer is lowered and centered manually. The extenso meter clip is

positioned carefully establishing a gauge length required for testing.

4. The series IX automatic materials testing software is used to handle the remaining

test. After entering the specimen and test parameters, the TEST icon from the

main menu is clicked to start the test.

5. The end of the test is marked by sample failure. At the end, max stress, breaking

strain (%), and 100% modulus will be printed out for each sample.

Tensile strength and ultimate elongation was measured and recorded at each test interval. This instrument is capable of collecting stress and strain data continuously during a test cycle. This data was used to calculate the crosslink densities, rate of deterioration of crosslink densities and the rate kinetics of the change, as presented in the chapter v.

26 CHAPTER IV

CONCEPTS OF RUBBER ELASTICITY

IV.1 Kinetic theory of Rubber elasticity

The behavior of rubbers when subjected to a static applied stress or strain shows that properties change significantly with both time of stressing or straining and with temperature. The implications of significant time and temperature dependence require that test conditions - rate, time, temperature and history of previous deformation - must be quoted if the results of a test are to be meaningful and applicable to the design of components.

The retractive force in stretched rubber is mainly due to entropy changes. Using this fact, it is possible to develop general expressions for the deformation of rubber units based upon the molecular structure of rubber. The derivations and the related background to the general expressions are commonly called the 'statistical theory', owing to the statistical treatment of the probable end-to-end distance of a typical single molecule; or alternatively the 'kinetic theory' owing to the analogy with the kinetic theory of gases, deriving from the consideration of thermal motions of molecular segments.

4 6 A typical rubber molecule consists of a long flexible chain of 10 - 10 units. With a large number of links the most probable end-to-end length is proportional to the square root of the number of links. This provides an explanation of the enormous extension

27 possible in soft rubber without rupture, since a chain of 10,000 units can elongate

10,00011 00 times in going from its probable to its maximum length.

Thus for large deformations, the stress-strain curve is predicted by equation IV.I.

cr = nkT(a-I/a,z) (lV.I)

Based on molecular (kinetic) theory, this equation predicts that a tensile stress-strain curve is nonlinear (Figure IV.I), with stress proportional to temperature. Smith (1962) further notes the same theory to develop equation (lV.2):

G = nkT = pRTI Me (lV.2) where Me equals molecular weight between cross links. This equation tells us an important fact that G (shear modulus) is directly related to Me. This equation does further show the dependence of modulus on absolute temperature T. The other symbols used are: n, the number of molecular chains per unit volume; p the density; k, Boltzmann's constant; and R the gas constant.

Summarizing equations IV.I and IV.2, in tum suggests the following equation IV.3:

(lV.3) where the original length, Lo , is increased to L (a = U Lo), and R T is the gas constant times the absolute temperature. The quantity n represents the number of active network chain segments per unit volume. The quantity 'n' is equal to pi Me, and is defined in the equation (lV.2).

The "molecular" basis, primarily using thermodynamics, originally was referred to as the "statistical theory" by Gauss, and later referred to as the "Gaussian theory" by

Kuhn (1936) and the "kinetic theory" by English researchers.

28 Strain (£)

Figure IV. 1 A stress-strain curve showing that equation IV. 1 is nonlinear (Hertz, 1991).

IV.2 Activation energy

The activation energy is the threshold energy, or the energy that must be overcome in order for a chemical reaction to occur. Activation energy may otherwise be denoted as the minimum energy necessary for a specific chemical reaction to occur. The activation energy of a reaction is usually denoted by Ea. It was Arrhenius who first suggested that the temperature dependence of the specific reaction rate k, could be correlated with activation energy by an equation of the type,

(IV.4) where k is the reaction rate constant; Ea the activation energy; R the gas constant; T the absolute temperature and A is a pre-exponential factor. Equation (IV.4), known as the

Arrhenius equation, has been verified empirically to give the temperature behavior of most reaction rate constants within experimental accuracy over fairly large temperature ranges.

29 Reaction

Product

Extent of Reaciion

Figure [V.2 Extent of reaction and the significance of activation energy

The activation energy Ea has been equated with a minimum energy that must be possessed by reacting molecules before the reaction will occur. From the kinetic theory of gases, the factor e -E~T gives the fraction of the collisions between molecules that together have this minimum energy E. The activation energy is determined experimentally by carrying out the reaction at several different temperatures. After taking the natural logarithm of equation (IV.4),

(IV.5)

it can be seen that a plot of In k versus 1fT should be a straight line whose slope is proportional to the activation energy.

-14.5 -,--~~~~-~~~~~~~ __ -14 ---- -13.5 -13 :;;; -12.5 .5 -12 -----""------11.5 -11 ------10.5

-10 +-~~---,-~~~,_____~~___,__~~---i 0.0036 0.0034 0.0032 0.003 0.0028 lIT

Figure IV.3 Sample Arrhenius plot in which the slope of the straight

line represents (-EIR)

30 CHAPTER V

PERFORMANCE RESULTS AND DISCUSSION

After aging, the dumbbell shaped coupons were pulled in the Instron Universal

Testing Machine with an extensometer to measure the tensile properties. In addition to ultimate stress and breaking strain, 100%, 300% and 500% modulus were also measured using the Instron. The measurements were taken on 0, 3, 6, 12, 20 and 30th days during the 30-day aging period. The ultimate stress and breaking strain values were then used to calculate crosslink density n, using Gaussian theory of elasticity.

V.I Gaussian Theory of Elasticity:

The Gaussian theory relating the stress-strain behavior of an elastomer to its crosslink density (from section IV.l) is given by,

(From equation IV.3)

Here, cr - Retractive Stress (Load / Initial Cross-sectional area)

a - Extension Ratio (Finallengthl Initial length)

n - Active network chain segments per unit volume (Crosslink Density)

RT - Gas constant times the absolute temperature.

The following plot was generated by the Instron tensile testing machine and it shows that the experimental data follows the theoretical behavior (as in figure IV.l) predicted by equation IV.3.

Retractive stress is calculated by dividing the stress/load obtained from Instron with the cross-sectional area (0.02 sq. in) of the elastomer dumbbell coupon and

31 Sample 10: 10-25-04-5-rt

50~r-----~----'-----~----~----~~~7'm-e-n:~1-r--~ Specimen: 2 -- ~mett3 40 ',/ .

/ ..c\f- 30 -'C CO .9 20

10

o 1 2 3 Displacement in

Figure V.I Plot generated by Instron which matches the trend infigure IV. I extension ratio is calculated from the strain data of the coupons with an initial length of

1.0 in. Crosslink density, n is calculated using the above values (step-wise calculations are presented in appendix II) and RT at three different temperatures and chloramine concentrations. The reproducibility of the data from the Instron was calculated prior to the experimentation and it is 0.06%. This error could be minimized by uniform clamping of the coupon with the jaws and calibrating the instrument before pulling a set of samples of each time interval.

Tables from V.l through V.6 exhibit the results obtained from the above calculations.

32 Crosslink Density, Mol/CC

Aging Time @ 22°C @45°C @ 70°C

o Days 2.46E-03 2.46E-03 2.46E-03

12 Days 2.60E-03 2.61E-03 2.58E-03

20 Days 2.56E-03 2.55E-03 2.57E-03

30 Days 2.43E-03 2.44E-03 2.06E-03

Table V.i Crosslink densities of EPDM-P, aged at ippm of chloramine concentration.

Crosslink Density, Mol/CC

Aging Time @ 22°C @45°C @ 70°C

o Days 2.46E-03 2.46E-03 2.46E-03

12 Days 2.45E-03 2. 17E-03 1.42E-03

20 Days 2.35E-03 1.87E-03 1.35E-03

30 Days 2.32E-03 1. 74E-03 1.28E-03

Table V.2 Crosslink densities of EPDM-P, aged at 30ppm of chloramine concentration.

Crosslink Density, Mol/CC

Aging Time @ 22°C @45°C @ 70°C

o Days 2.46E-03 2.46E-03 2.46E-03

12 Days 2.43E-03 2.09E-03 1.44E-03

20 Days 2.41E-03 1. 89E-03 1.20E-03

30 Days 2.28E-03 1.63E-03 1.15E-03

Table V.3 Crosslink densities of EPDM-P, aged at 60ppm of chloramine concentration.

33 Crosslink Density, Mol/CC

Aging Time @ 22°C @4SoC @ 70°C

o Days l.77E-03 l.77E-03 1.77E-03

12 Days 1.S6E-03 1.93E-03 2.l2E-03

20 Days 1.94E-03 2.03E-03 2.38E-03

30 Days 1.99E-03 2.llE-03 2.4SE-03

Table V.4 Crosslink densities of EPDM-S, aged at Jppm of chloramine concentration.

Crosslink Density, MollCC

Aging Time @ 22°C @4SoC @ 70°C

o Days 1.8SE-03 1.SSE-03 1.8SE-03

12 Days 1.83E-03 1.8lE-03 1.8SE-03

20 Days l.82E-03 1.77E-03 1.72E-03

30 Days 1. 83E-03 1.70E-03 1.60E-03

Table V.5 Crosslink densities of EPDM-S, aged at 30ppm of chloramine concentration.

Crosslink Density, Mol/CC

Aging Time @ 22°C @4SoC @ 70°C

o Days 1.77E-03 1.77E-03 1.77E-03

12 Days 1.77E-03 1.70E-03 1.66E-03

20 Days 1.74E-03 l.6SE-03 l.49E-03

30 Days 1.73E-03 1.S6E-03 1.43E-03

Table V.6 Crosslink densities of EPDM-S, aged at 60ppm of chloramine concentration.

34 Each of these calculated CLD values is the average of three CLDs of the three

tensile samples. Standard deviations of these values are calculated and an example table

5 is presented below. All the standard deviations are small, in the order of 10- .

Crosslink Densit)', Mol/CC

Aging Time @22°C @45°C @70°C

oDays 2.46E-03 (+/- 1.8 IE-05) 2.46E-03 (+/- 1.81 E-05) 2.46E-03 (+/- 1.8 IE-05)

12 Days 2.43E-03 (+/- 4.55E-05) 2.09E-03 (+/- 9.67E-OS) J .44E-03 (+/- 3.38E-05)

20 Days 2.41 E-03 (+/- 7.65E-(5) 1.89E-03 (+/- 6.66E-(5) 1.20E-03 (+/- 7.14E-05)

30 Days 2.28E-03 (+/- 1.50E-(5) 1.63E-03 (+/- 7. I3E-(5) 1.ISE-03 (+/- 232E-OS)

Table V.6 (a) Standard deviations of the respective CLDsfor EPDM - P aged at 60ppm

of chloramine concentration.

Experimental data obtained from Instron is checked for its fit to the Gaussian

theory. Based on the theory that the stress (cr) is linearly proportional to the factor (a - . 2 V( ), experimental data is checked for linearity. Figure V.l (a) is an example of the

linearity check of Instron data. Every data cycle exhibited slight deviation in the initial

stage due to the slack of dumbbell coupon and adjustment of speed of the moving jaws.

V.2 Rate of change of crosslink densities

The structure of a cross-linked elastomer may be idealized as the molecular

structure in section 11.7.1. The primary chains are cross-linked at many points along their

length. The implication of a crosslink density of l.77E-03 Mol/CC is that, there are 20-30 cross-links per primary molecule of molecular weight in the order of 1 x 105 glmol. [The

CLD values here are slightly higher than the values in literature, -E-04 MollCC because

these are pipe gaskets and seals and, as such, are harder than the rubber for other

applications] .

35 Fit of the data to the Gaussian theory BOOE+07.,------.,

7.o0E+07 +------::JIIf=---!

6.ooE+07 +------~~-----'

5.o0E+07 +------~=------i

4.o0E+07 +------211''''------' a 3.o0E+07 +------,...,.------!

2.0oE+07 +-----~~------j

1.ooE+07 +----,.,...tS,,,£------i

o .DO E+ 00 +fr-,...,.,...,.,...,..,..r'T'T'T''I''T'T''I'''''''''''l''''I"TT'''''''''''M"M''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''T''I''T'1'''''''''''I"T'T'T'TTT''1''TT''i 0.01 0.36 0.62 0.B4 1.02 1.18 1.33 1.4B 1.61 174

a-lIa2

Figure V.I (a) A plot of experimental data from Instron showing the linearity of the data

The crosslink densities (CLD) decreased over time and with increasing temperature during their aging. The rate of change of the property is calculated by determining the change in CLD over the time of aging (calculations are explained in appendix II). Tables

V.7 and V.8 exhibit the rates calculated for EPDM-P and EPDM-S respectively.

Chloramine Rate of change of CLD, 6CLDI Day Concentration PPM @ 22°C @4SoC @70°C

1 -8.919E-07 -1.01E-06 -1.1729E-OS

30 -S.lSSE-06 -2.49E-OS -3.823E-OS

60 -S.784E-06 -2.74E-OS -4.3639E-OS

Table V.7 Rate of change of crosslink densities in 30 days for EPDM-P

36 Chloramine Rate of change of CLD, 6CLDI Day I Concentration PPM @ 22°C 35 °C @ 70°C

I 7.497E-06 1. I 64E-05 2.3493E-05

30 -4.7ISE-07 -4.977E-06 -S.60SE-06

60 -1.44SE-06 -6.976E-06 -1.209E-05

Table V.8 Rate of change of crosslink densities in 30 days for EPDM-S

The loss in crosslink densities of sulfur and peroxide cured EPDM rubber during chloramine aging followed common patterns. The rates of decrease were largest in the early stages of aging. To understand the underlying mechanism of deterioration, the rates are studied with respect to chloramine concentration at three temperatures. The plots of rate of change of CLD with respect to concentration are shown below.

Chloramine Concentration PPM o 30 60 o __ :-:- ______..... ------+------9 -0.00001 +... ~------'---c;-----.------____I u .... , 4-< ~ -0.00002 ---- ~ " , " ...... -t. ~ -0.00003 +------'---=c-,------J o II) ~ -0.00004 +------=."'=------:' -~..--;.=--.-_-. -_-. _-. _--. _-. ------.

-0.00005 --'--______--1

- ~ - 22C ...... 45C - .•. - 70C

Chart V.l Rate of change of CLD of EPDM - P with respect to chloramine

concentration at three temperatures.

37 At 22°C, the deterioration rate is very low with a rate of 8.9 E-07 (mol/cm3)/day or

6CLD / Day. At 1ppm, the rates at 22°C and 45°C are very close when compared to the rate at 70°C i.e., 1.1729E-05 mollcm3/day. This trend exemplifies the effect of temperature on degradation of rubber properties. The 45°C and 70°C rate curves followed the same pattern; a large increase in the rate at low concentrations and a constant high rate at higher concentrations.

Chloramine Concentration PPM o 30 60 0.00003 -,------,.. ---...... ------0.00002 ... , ". , 0.00001 +-----'-"-.--,--.. -- -- ~ ..,------... -- ...... --.: '':''-:''- ~:.:':-:;--. ---- O+------~~.. ~~~~~~---_-_-_-_-_-_-_-_-_-_-_-_-_~ ' ...... ::::: -.•...... -0.00001 +------~~~--- -- .. -.- ... _------

-0.00002 ...1.--______----1

- -+- - 22 ...•... 45 - .•. - 70

Chart V.2 Rate of change of CLD of EPDM - S with respect to chloramine concentration

at three temperatures

At 1ppm, EPDM-S exhibited hardening or an increase in CLD, rather than a decrease in the property due to degradation. This could be due to formation of additional crosslinks within the molecule. According to the structure of vulcanized rubber, a link of two or more sulfur atoms are attached to the main chain. These extra sulfur atoms could break up with the main link and form additional links to other parts of the molecular chain. This process can result in the increase in CLD at lower concentrations, where the

attacking chloramine is not sufficient enough to break the additional crosslinks. The

38 resulting hardening of the material was evident in other properties as well. Hardness was increased throughout the period of aging. This data is presented below in the chart V.3.

At high concentrations, the pattern of the rate of deterioration was similar to that of

EPDM-P. The rate flattened at these high concentrations between 30ppm and 60ppm of chloramine. The greatest extent of hardening (at Ippm) as well as deterioration (at

60ppm) occurred at 70°C.

Shore A Hardness - EPDM Sulfur

10~------'------'

_- ... - 70C S +------~~---~:-~---; -- --¥ ---

._ ..... _._ ..... ·145C 4 , , ~ --.­ .... I I,.' ;>--. .. -+-. ..,. . 2 +-~/'--. '~', ,... """"------'--~ •.•'-"'.~:.-c-:_-:---. -_-. --,-;..-,: ..._~. -_-. _---::--T"'-----"~~=-=-='--'--""~~~"'"'.-=-=-=. -:-.----1 RT /,"/ ...... 4';- o It o 5 10 15 20 25 30 Days -·.·-RT ...•... 45C - ...... -70C

Chart V.3 Percent change of hardness of EPDM - S during the 3D-day aging period at

1ppm chloramine concentration and three temperatures.

V. 3 Model of rate kinetics

Rate expressions were developed for both the EPDM rubbers to describe the observed deterioration of crosslink densities based on the data presented in tables V.7 and

V.S. For EPDM - P, it is expressed in the equation below:

(V.2) Rate = kIk2[C] kI[C]+ k2

39 Or, 1 Rate=---- 1 1 --+- (V.3) kl[C] b The above rate equation suggests that the aging reaction is a two step process. At lower concentrations, the rate of reaction is a first order function of chloramine concentration, i.e. kIfC}. At higher concentrations, the rate approaches zero order with a rate of k2. It is understood that at these high concentrations the amount of chloramine present is abundant and the rate is no longer dependent on the concentration.

Quantitatively, from equation V.3, when the value of kIfC} is high, (1 / kIfC]) approaches zero and rate will be equal to k2• Therefore, the course of reaction could be explained as a series of two reactions with kJ being the rate constant for the first reaction and k2 for the second reaction. When the rates at all the three temperatures are extrapolated to infinite concentration, the rates are equal to k2 which are closer to the rates at 60ppm suggesting that rate approaches a constant value and becomes a zero order reaction. The values are presented in the table V.9 below:

Chloramine Rate of change of CLD, 6CLDI Day Concentration PPM @22°C @4SoC @70°C

60 -S.78E-06 -2.74E-OS -4.36E-OS

Infinity -6.3SE-06 -3.41E-OS -4.40E-OS

Table V.9 Extrapolated reaction rates compared with those of60ppmfor EPDM - P

The developed model is checked for a best fit of data at all the three temperatures using nonlinear regression. The resulting r2 values at 22°C, 4SoC and 70°C were 0.99,

0.99 and 0.98 respectively. The rate constants in the equation were also estimated by this regression and are presented in the table V.1 0 below:

40 Rate Constants Temperature kJ k2

22°C -9.8IE-07 -6.35E-06

45°C -2.68E-06 -3.4IE-05

70°C -1.53E-05 -4.40E-05

Table V.lO Rate constants of EPDM - P estimated by nonlinear regression

Deterioration of EPDM-S crosslink densities suggests a similar mechanism except the increase in CLD at lppm, as explained in section V.2. This increasing trend in crosslink densities at 1ppm introduces a new constant in the rate equation. The model is depicted in the equation below:

klk2[C] (V.4) Rate = b+ [] kl C +k2

Or, 1 Rate = b + ---­ I 1 --+- (V.5) kJ[C] b

EPDM - S exhibited first order increase in the rate at lower concentrations, like

EPDM - P. This change is quantified by the term kIfC] shown above in the equation V.5.

At higher concentrations the rate was high but approached a constant value, behaving like a zero order reaction. The positive change in crosslink densities at 1ppm is quantified using a constant k3• The increase in CLD pertains to the formation of additional cross links within the molecule. This increase is a zero order change with no dependence on chloramine concentration but depending on temperature.

41 The model was verified for best fit using nonlinear regression and the r2 values are

0.999 at all the three temperatures. The three rate constants were also estimated by regression and are tabulated in the table V.ll. Extrapolated rates were calculated with concentration approaching infinity. At infinity, the rates are equal to the sum of k2 and k3 at their respective temperatures. These values are close to the rates at 60ppm proving that the curve straightens and ultimately approaches a constant. These ultimate constants are presented in the table V.12 for comparison. In a physical point of view, kJ represents the rate of loss of CLD per day and k2 represents the rate of loss per day and per PPM of chloramine in the solution. Rate constants include everything that affects the reaction rate other than concentration, mainly temperature.

Rate Constants Temperature kJ k2 k3

22°C -1.68E-06 -1.16E-OS 8.96E-06

4SoC -3.S8E-06 -2.42E-OS 1.48E-OS

70°C -7.86E-06 -4.6SE-OS 3.02E-OS

Table V.ll Rate constants of EPDM - S estimated by regression

Chloramine Rate of change of CLD, 6CLDI Day Concentration PPM @ 22°C @4SoC @ 70°C

60 -1.448E-06 -6.976E-06 -1.209E-OS

Infinity -2.6S E-06 -9.43 E-06 -1.63 E-OS

Table V.l2 Extrapolated reaction rates compared with those of60ppmfor EPDM - S

42 V.4 Temperature dependence of rate and activation energies

Arrhenius related the variation in reaction rate to temperature in a relationship as

shown in equation V.6:

(V.6)

where k is the reaction rate constant; Ea the activation energy; R the gas constant; T the

absolute temperature and A is a pre-exponential factor. From this expression, the natural logarithm of reaction rate k is proportional to 1 / T, and the slope is equal to -Ea / R. As reaction rates at three temperatures are known, the activation energies, Ea were calculated for both the materials.

A modified version of the equation V.6 is used to predict the degradation rates and

service lives at operating temperatures. The modified equation is presented as follows:

(V.7) ~= Aexp[-~(~ __l )l Y2 RT Tl T2 J where rJ is the reaction rate at temperature TJ etc., A is a constant and Ea is the activation energy.

At three temperatures 22°C, 45°C and 70°C, aging of EPDM - P has two reaction rate constants at each temperature and EPDM - S has three rate constants. Therefore, two activation energies are calculated for the two reaction steps for EPDM - P and three activation energies for EPDM -So A linear regression of the Arrhenius equation in the form V.6 yielded the activation energy for a low chloramine concentration aging, EaJ =

48 kJ/mol and for high concentration aging, Ea2 = 34 kJ/mol for EPDM - P. An Arrhenius

43 plot of EPDM - P rate constants of the low concentration process vs. 1 I T is presented in the chart V.4 as follows:

Arrhenius Plot for EPDM - P

-14.5 -14 -13.5 -13 ...... ~ -12.5 ..E -12 -11.5 -11 -10.5 ------

-10+------~------,------,------~ 0.0036 0.0034 0.0032 0.003 0.0028 1 IT

Chart VA Arrhenius plot of low concentration reaction rate constants vs. lIT for

EPDM-P

For EPDM - S, EaJ = 27 kllmol, Ea2 = 24 kllmol and the activation energy for the process of additional cross linking at 1ppm, Ea3 = 21 kllmol. All the Ea values, pre- exponential factors and R2 values are tabulated in the tables V.l3 and V.14 for EPDM - P

and EPDM - S respectively. The low activation energies of EPDM - S suggests that

EPDM - P has better resistance for chloramine aging than EPDM - S. These activation energies of EPDM rubber are slightly lower than the values in literature (6, 7) for similar type of rubber compounds obtained by various methods. Activation energies of Natural rubber compounds for aging in sea water were in the range of 55 - 65 kllmol. Analyses based on air aging yielded higher values, in the range of 86 - 94 kllmol. During heat

44 aging of EPDM membranes from 116°C to 150°C, activation energies for property

changes ranged from 89 - 102 kllmol.

The following chart is the Arrhenius plot for the low concentration reaction process

for EPDM - s:

Arrhenius Plot for EPDM - S -13.5

-13 ~ --

-12.5 ~

- -12 ~ ~ -..., ..s -11.5 -11

-10.5 I

-10 I I I I 0.0036 0.0034 0.0032 0.003 0.0028 liT

Chart V.S Arrhenius plot of low concentration reaction rate constants vs. 1fTfor

EPDM-S

For EPDM - S, linear regression of In k vs. liT yielded the activation energies with high R2 values of 0.98 to 0.99 when compared to EPDM - P for which the R2 values are

0.86 and 0.97. The frequency factors or pre-exponential factors are higher for EPDM- P than EPDM - S (tables V.13 and V.14). A value of 264.9 for the frequency factor of

EPDM - P indicates that the number of collisions between the reactants which have the correct orientation to form products is high when compared to 0.102 for EPDM - S. Also, the frequency factor of the first reaction step in the series is higher than the one for

second step which suggests that at lower concentrations, the collision of reactants is

greater than at higher concentrations.

45 The activation energies (Ea), frequency factors (A) and the respective r2 values of

both the materials are reported below in the table V.13 and V.14:

Ea KJ/Mol A R2 kJ 47.9 264 0.97 k2 34.2 8.92 0.86

Table V.13 The Ea, A and R2 values of both the reaction steps for EPDM -Po

Ea KJ/Mol A R2 kJ 27.1 0.10 1.00 k2 24.3 0.24 1.00 k3 21.2 0.05 0.98

Table V.14 The Ea, A and R2 values of the reaction steps for EPDM -So

V.S Prediction of service life

With the help of activation energies and rate constants, service life or the time taken

for the material to lose a particular amount of crosslink density can be determined. A

sample calculation is presented below:

Material: EPDM-P

Service conditions: Temperature - 24C, Chloramine concentration -lppm

% Deterioration in CLD at which the field part fails - 60%

Initial CLD: 2.46 x 10-3 Mol/cm3

With T = 297 K (24C) kJ =9.80E-07, k2 = 8.68E-06

46 With [C] = 1ppm

Rate = 8.81E-07 mol/cm3.day

Life = (0.6 * 2.46E-03) I (8.81E-07)

= 1476 Days =4 years

This service time represents a material which is fully and continuously exposed to chloramine unlike the field parts. Also the ratio of exposed surface area to volume of experimental coupon is very high when compared to the field parts, which gives a shorter life span for the laboratory samples. Using a geometric factor of the elastomeric part shape and surface area, a more precise service life could be calculated. Also, the same kind of calculation cannot be performed for EPDM-S due to the positive change in CLD at concentrations below 15ppm.

V.6 Conclusions

The calculations and rate equations developed have provided useful insights into the degradation mode and mechanism of chloramine induced EPDM failure. The key conclusion here is that, the degradation process is a two step series reaction. The first step of the series has its reaction rate proportional to the chloramine concentration and also has higher activation energy (48 KllMol vs. 34 KllMol for EPDM -P and 27 KllMol vs. 24 KllMol for EPDM - S) than the second step for both the EPDMs. The second reaction which approaches a zero order reaction state has lower activation energy than the first reaction, suggesting that the first reaction is the limiting step in the aging process.

The activation energies of EPDM - P are almost double the activation energies of

EPDM - S, which explains the reason behind the greater resistance of EPDM - P towards chloramine attack than EPDM - S, especially at lower concentrations. Additionally, these

47 activation energies explain the low changes in other properties like swelling and hardness

in EPDM - P when compared to EPDM - S or other rubber materials.

Accelerated aging experiments can provide useful insights into the agmg

phenomenon in a feasible time scale. Modeling of the change in crosslink density data

and applying Arrhenius equation to the reaction rates aids in understanding the failure

modes at a molecular level.

V.7 Recommendations

Based on the study conducted so far and the results obtained, the following

recommendations were made to further investigate elastomer degradation:

1. A similar kind of calculations can be performed for SBR, Nitrile rubber, Natural

rubber and Neoprene as well, to understand their failure modes as every rubber

is compounded in a different way.

2. Insufficient data was the limitation for every similar study conducted in the past.

Though there were three temperatures for the Arrhenius plot, another point of

study between room temperature and 45°C would reinforce the results.

3. Aging of EPDM - S at Ippm could be carried for a longer'time, which is for an

additional 60 days, to understand the extent of additional cross linking.

48 REFERENCES

1. Class, Jay B., "A review of the fundamentals of cross linking with peroxides", Jour.

Rubber world, August 1999.

2. Hertz, Daniel L. Jr., "An analysis of rubber under strain from an engineering

perspective", Jour. Elastomerics, December 1991.

3. John A. Dean, Lange's Handbook of Chemistry, Fifteenth Edition, McGraw - Hill

Inc.

4. Maurice Morton, Rubber Technology, Second Edition.

5. Mitchell, Julian M., "EP Polymer Selection and compound considerations for

chloramine resistance", Rubber World, June 1999.

6. Mott P.H, Roland C.M., "Aging of Natural rubber in air and seawater", Jour. Rubber

Chemistry and Technology, March 2001.

7. Paeglis, Amus U., "A simple model for predicting heat aging of EPDM Rubber",

Jour. Rubber Chemistry and Technology, June 2004.

8. Reiber, Steve, "Investigating the Effects of Chloramine on elastomer degradation",

Jour. AWWA, 1993.

9. Richard, Bonds W., "Effect of chloramines on ductile-iron pipe gaskets of various

elastomer compounds", Jour. AWWA, April 2004.

10. Simmons C. L., Evanson P.P., "Effects of additives in domestic water systems on

rubber vulcanizates", proceedings - Rubber Division, ACS, Dallas, 1988.

49 11. Smith, T.L. 1962. "Stress-Strain- Time-Temperature relationships for Polymers."

ASTM Special Technical Publication No.325.

12. Vicic John c., Cain David, Manuel Maligas, "Estimating elastomer failure life using

environmental tests", Jour. Elastomerics, May 1992.

50 APPENDIX I

NOMENCLATURE

Activation Energy

A Frequency factor or Pre-exponential factor

EPDM - S Ethylene Propylene Diene Methylene rubber - sulfur cured

EPDM - P Ethylene Propylene Diene Methylene rubber - peroxide cured k Reaction rate constant

D Dilution of solution for titration sample for chlorine titrimeter

V Sample used for titration in ml

A Volume of phenylarsine oxide solution required for titration in ml. cr Retractive Stress (Load / Initial Cross-sectional area) a Extension Ratio (Finallengthl Initial length) n orCLD Active network chain segments per unit volume (Crosslink

Density,)

R Universal gas constant

T Absolute temperature in oK

[C) Chloramine concentration in ppm

51 Appendix II

SAMPLE CALCULATIONS

I. Data from Instron:

Data generated by the Instron tensile testing machine is presented in column 1 as

'Displacement' and in column 4 as 'Stress'.

Displacement L =Dis + Stress (mm) 2S.4 a =L/2S.4 Stress (Mpa) (dynes/cm2) a - a "2 0.3146 25.7146 1.01 0.7775 7.77E+06 0.04 0.7264 26.1264 1.03 0.8402 8.40E+06 0.08 1.1290 26.5290 1.04 0.8964 8.96E+06 0.13 1.5311 26.9311 1.06 0.9500 9.50E+06 0.17 1.9615 27.3615 1.08 1.0012 1.00E+07 0.22 2.3935 27.7935 1.09 1.0497 1.05E+07 0.26 2.8229 28.2229 1.11 1.1009 1.10E+07 0.30 3.2381 28.6381 1.13 1.1485 1.15E+07 0.34 3.6600 29.0600 1.14 1.1909 1.19E+07 0.38 4.0705 29.4705 1.16 1.2349 1.23E+07 0.42 4.5134 29.9134 1.18 1.2750 1.27E+07 0.46 4.9238 30.3238 1.19 1.3184 1.32E+07 0.49 5.3480 30.7480 1.21 1.3602 1.36E+07 0.53 5.7693 31.1693 1.23 1.4033 1.40E+07 0.56 6.1826 31.5826 1.24 1.4496 1.45E+07 0.60 6.6080 32.0080 1.26 1.4880 1.49E+07 0.63 7.0107 32.4107 1.28 1.5334 1.53E+07 0.66 7.4013 32.8013 1.29 1.5715 1.57E+07 0.69 7.8114 33.2114 1.31 1.6133 1.61E+07 0.72 8.2182 33.6182 1.32 1.6547 1.65E+07 0.75 8.6005 34.0005 1.34 1.6981 1.70E+07 0.78 8.9967 34.3967 1.35 1.7443 1. 74E+07 0.81 9.3859 34.7859 1.37 1.7899 1.79E+07 0.84 9.7714 35.1714 1.38 1.8370 1.84E+07 0.86

52 10.1457 35.5457 1.40 1.8785 1. 88E+07 0.89 10.5485 35.9485 1.42 1.9200 1.92E+07 0.92 10.9259 36.3259 1.43 1.9654 1.97E+07 0.94 11.2816 36.6816 1.44 2.0155 2.02E+07 0.96 11.6482 37.0482 1.46 2.0663 2.07E+07 0.99 11.9966 37.3966 1.47 2.1183 2. 12E+07 1.01 12.3568 37.7568 1.49 2.1699 2. 17E+07 1.03 12.7250 38.1250 1.50 2.2193 2.22E+07 1.06 13.0807 38.4807 1.51 2.2713 2.27E+07 1.08 13.4343 38.8343 1.53 2.3242 2.32E+07 1.10 13.7964 39.1964 1.54 2.3753 2.38E+07 1.12 14.1394 39.5394 1.56 2.4330 2.43E+07 1.14 14.4887 39.8887 1.57 2.4906 2.49E+07 1.16 14.8380 40.2380 1.58 2.5491 2.55E+07 1.19 15.1930 40.5930 1.60 2.6063 2.61E+07 1.21 15.5381 40.9381 1.61 2.6649 2.66E+07 1.23 15.8671 41.2671 1.62 2.7281 2.73E+07 1.25 16.1968 41.5968 1.64 2.7892 2.79E+07 1.26 16.5347 41.9347 1.65 2.8489 2.85E+07 1.28 16.8653 42.2653 1.66 2.9112 2.91E+07 1.30 17.1951 42.5951 1.68 2.9775 2.98E+07 1.32 17.5118 42.9118 1.69 3.0429 3.04E+07 1.34 17.8558 43.2558 1.70 3.1095 3. 11E+07 1.36 18.1679 43.5679 1.72 3.1758 3. 18E+07 1.38 18.4744 43.8744 1.73 3.2439 3.24E+07 1.39 18.7820 44.1820 1.74 3.3095 3.31E+07 1.41 19.1156 44.5156 1.75 3.3801 3.38E+07 1.43 19.4241 44.8241 1.76 3.4543 3.45E+07 1.44 19.7357 45.1357 1.78 3.5250 3.53E+07 1.46 20.0614 45.4614 1.79 3.5934 3.59E+07 1.48 20.3649 45.7649 1.80 3.6652 3.67E+07 1.49 20.6696 46.0696 1.81 3.7413 3.74E+07 1.51 20.9909 46.3909 1.83 3.8114 3.81E+07 1.53 21.2811 46.6811 1.84 3.8878 3.89E+07 1.54 21.6043 47.0043 1.85 3.9655 3.97E+07 1.56 21.9118 47.3118 1.86 4.0454 4.05E+07 1.57 22.2105 47.6105 1.87 4.1237 4. 12E+07 1.59 22.5120 47.9120 1.89 4.2003 4.20E+07 1.61 22.8273 48.2273 1.90 4.2762 4.28E+07 1.62

53 23.1239 48.5239 1.91 4.3589 4.36E+07 1.64 23.4177 48.8177 1.92 4.4382 4.44E+07 1.65 23.7191 49.1191 1.93 4.5176 4.52E+07 1.67 24.0061 49.4061 1.95 4.6002 4.60E+07 1.68 24.2912 49.6912 1.96 4.6873 4.69E+07 1.70 24.6005 50.0005 1.97 4.7696 4.77E+07 1.71 24.8967 50.2967 1.98 4.8473 4.85E+07 1.73 25.2073 50.6073 1.99 4.9342 4.93E+07 1.74 25.5069 50.9069 2.00 5.0187 5.02E+07 1.76 25.8008 51.2008 2.02 5.1031 5.lOE+07 1.77 26.0831 51.4831 2.03 5.1966 5.20E+07 1.78 26.3965 51.7965 2.04 5.2787 5.28E+07 1.80 26.6755 52.0755 2.05 5.3647 5.36E+07 1.81 26.9740 52.3740 2.06 5.4542 5.45E+07 1.83 27.2624 52.6624 2.07 5.5456 5.55E+07 1.84 27.5470 52.9470 2.08 5.6339 5.63E+07 1.85 27.8310 53.2310 2.10 5.7211 5.72E+07 1.87 28.1142 53.5142 2.11 5.8086 5.81E+07 1.88 28.3962 53.7962 2.12 5.8998 5.90E+07 1.90 28.6737 54.0737 2.13 5.9951 6.00E+07 1.91 28.9544 54.3544 2.14 6.0824 6.08E+07 1.92 29.2305 54.6305 2.15 6.1755 6. 18E+07 1.93 29.5246 54.9246 2.16 6.2662 6.27E+07 1.95 29.8128 55.2128 2.17 6.3638 6.36E+07 1.96 30.0958 55.4958 2.18 6.4594 6.46E+07 1.98 30.3667 55.7667 2.20 6.5466 6.55E+07 1.99 30.6552 56.0552 2.21 6.6413 6.64E+07 2.00 30.9379 56.3379 2.22 6.7394 6.74E+07 2.01 31.2172 56.6172 2.23 6.8369 6.84E+07 2.03 31.5139 56.9139 2.24 6.9313 6.93E+07 2.04 31.7994 57.1994 2.25 7.0254 7.03E+07 2.05 32.0889 57.4889 2.26 7.1230 7. 12E+07 2.07 32.3834 57.7834 2.27 7.2239 7.22E+07 2.08 32.6729 58.0729 2.29 7.3188 7.32E+07 2.10 32.9842 58.3842 2.30 7..4129 7.41E+07 2.11 33.2678 58.6678 2.31 7.5140 7.51E+07 2.12 33.5610 58.9610 2.32 7.6155 7.62E+07 2.14 33.8579 59.2579 2.33 7.7160 7.72E+07 2.15 34.1493 59.5493 2.34 7.8102 7.81E+07 2.16

54 34.4526 59.8526 2.36 7.9041 7.90E+07 2.18 34.7321 60.1321 2.37 8.0100 8.01E+07 2.19 35.0394 60.4394 2.38 8.1046 8.lOE+07 2.20 35.3195 60.7195 2.39 8.2037 8.20E+07 2.22 35.6196 61.0196 2.40 8.3037 8.30E+07 2.23 35.9055 61.3055 2.41 8.4047 8.40E+07 2.24 36.1799 61.5799 2.42 8.5002 8.50E+07 2.25 36.4718 61.8718 2.44 8.6014 8.60E+07 2.27 36.7625 62.1625 2.45 8.7018 8.70E+07 2.28 37.0401 62.4401 2.46 8.8054 8.81E+07 2.29 37.3294 62.7294 2.47 8.8999 8.90E+07 2.31 37.6204 63.0204 2.48 9.0028 9.00E+07 2.32 37.9336 63.3336 2.49 9.1058 9. 11E+07 2.33 38.2209 63.6209 2.50 9.2017 9.20E+07 2.35 38.4935 63.8935 2.52 9.3020 9.30E+07 2.36 38.7976 64.1976 2.53 9.4096 9.41E+07 2.37 39.0981 64.4981 2.54 9.5068 9.51E+07 2.38 39.3958 64.7958 2.55 9.6096 9.61E+07 2.40 39.7101 65.1101 2.56 9.7113 9.71E+07 2.41 39.9949 65.3949 2.57 9.8181 9.82E+07 2.42 40.3017 65.7017 2.59 9.9227 9.92E+07 2.44 40.6045 66.0045 2.60 10.0240 1.00E+08 2.45 40.9046 66.3046 2.61 10.1288 1.01E+08 2.46 41.2062 66.6062 2.62 10.2297 1.02E+08 2.48 41.5189 66.9189 2.63 10.3332 1.03E+08 2.49 41.8277 67.2277 2.65 10.4366 1.04E+08 2.50 42.1439 67.5439 2.66 10.5428 1.05E+08 2.52 42.4626 67.8626 2.67 10.6398 1.06E+08 2.53 42.7828 68.1828 2.68 10.7488 1.07E+08 2.55 43.0843 68.4843 2.70 10.8522 1.09E+08 2.56 43.3984 68.7984 2.71 10.9570 1.10E+08 2.57 43.7156 69.1156 2.72 11.0591 1.11E+08 2.59 44.0330 69.4330 2.73 11.1669 1. 12E+08 2.60 44.3503 69.7503 2.75 11.2779 1. 13E+08 2.61 44.6482 70.0482 2.76 11.3906 1. 14E+08 2.63 44.9271 70.3271 2.77 11.5092 1. 15E+08 2.64 45.2162 70.6162 2.78 11.6305 1. 16E+08 2.65 45.4984 70.8984 2.79 11.7651 1.18E+08 2.66 45.7513 71.1513 2.80 11.9012 1.19E+08 2.67

55 46.0124 71.4124 2.81 12.0473 1.20E+08 2.69 46.1992 71.5992 2.82 12.1850 1.22E+08 2.69 46.4531 71.8531 2.83 12.3313 1.23E+08 2.70 46.6991 72.0991 2.84 12.4831 1.25E+08 2.71 46.9907 72.3907 2.85 12.6333 1.26E+08 2.73 47.2345 72.6345 2.86 12.7832 1.28E+08 2.74 47.4579 72.8579 2.87 12.9351 1.29E+08 2.75 47.7023 73.1023 2.88 13.0893 1.31E+08 2.76 47.9424 73.3424 2.89 13.2451 1.32E+08 2.77 48.1754 73.5754 2.90 13.3959 1.34E+08 2.78 48.4255 73.8255 2.91 13.5532 1.36E+08 2.79 48.6673 74.0673 2.92 13.7004 1.37E+08 2.80 48.9252 74.3252 2.93 13.8539 1.39E+08 2.81 49.1855 74.5855 2.94 14.0054 1.40E+08 2.82 49.4512 74.8512 2.95 14.1513 1.42E+08 2.83 49.7160 75.1160 2.96 14.3022 1.43E+08 2.84 49.9718 75.3718 2.97 14.4484 1.44E+08 2.85 50.2587 75.6587 2.98 14.5934 1.46E+08 2.87 50.5422 75.9422 2.99 14.7351 1.47E+08 2.88 50.8239 76.2239 3.00 14.8722 1.49E+08 2.89 51.1199 76.5199 3.01 15.0161 1.50E+08 2.90 51.4026 76.8026 3.02 15.1566 1.52E+08 2.91 51.7053 77.1053 3.04 15.3000 1.53E+08 2.93 51.9926 77.3926 3.05 15.4372 1.54E+08 2.94 52.2978 77.6978 3.06 15.5738 1.56E+08 2.95 52.6095 78.0095 3.07 15.7090 1.57E+08 2.97 52.9221 78.3221 3.08 15.8381 1.58E+08 2.98 53.2387 78.6387 3.10 15.9775 1.60E+08 2.99 53.5586 78.9586 3.11 16.1105 1.61E+08 3.01 53.8875 79.2875 3.12 16.2435 1.62E+08 3.02 54.2142 79.6142 3.13 16.3754 1.64E+08 3.03 54.5298 79.9298 3.15 16.5055 1.65E+08 3.05 54.8659 80.2659 3.16 16.6299 1.66E+08 3.06 55.1869 80.5869 3.17 16.7584 1.68E+08 3.07 55.5086 80.9086 3.19 16.8861 1.69E+08 3.09 55.8540 81.2540 3.20 17.0178 1.70E+08 3.10 56.1807 81.5807 3.21 17.1385 1.71E+08 3.11 56.5028 81.9028 3.22 17.2669" 1.73E+08 3.13 56.8423 82.2423 3.24 17.3877 1.74E+08 3.14

56 57.1764 82.5764 3.25 17.5148 1.75E+08 3.16 57.4987 82.8987 3.26 17.6303 1.76E+08 3.17 57.8405 83.2405 3.28 17.7555 1.78E+08 3.18 58.1700 83.5700 3.29 17.8786 1.79E+08 3.20 58.5088 83.9088 3.30 17.9988 1.80E+08 3.21 58.8405 84.2405 3.32 18.1186 1.81E+08 3.23 59.1738 84.5738 3.33 18.2349 1.82E+08 3.24

Column 2: Final length of the coupon = Strained length + Initial length (25.4 mm)

E.g. L = 0.3146 + 25.4 = 25.7146 mm.

Column 3: Extension ratio, a = Final length I Initial Length

E.g. a = 25.7146 I 25.4 = 1.01

Column 5: Converting Stress (cr) in MPa to Dynes/Cm2 by mUltiplying with 107

E.g. 0.775 MPa = 7.75 x 106 Dynes/Cm2

2 Column 6: (a - II a ) is calculated,

E.g. (1.01 - 111.01 2) = 0.04

II. Calculating Cross link density, N:

2 2 From the Gaussian theory, cr = NRT (a - II a ), slope of the line cr vs. (a - II a ) is equal to NRT, from which N is determined by dividing with RT.

2 2 E.g. Slope of cr vs. (a - 11 a ) = 5.96E+07 Dynes/Cm

5.96E + 07 = 2.43E-03 Mol/Cm3 (8.31E + 07)X 295

III. Calculation of rate of change in CLD:

At a particular concentration, CLDs at all the three temperatures and times are calculated. As a sample, the CLDs of EPDM - P at 22°C for all the four test days are presented below:

57 Time(Days) @22°C @45°C @70°C 0 2.46E-03 2.46E-03 2.46E-03 12 2.43E-03 2.09E-03 1.44E-03 20 2.41E-03 1. 89E-03 1.20E-03 30 2.28E-03 1.63E-03 1.15E-03

At each of the three temperatures, rate of change in CLDs is equal to the slopes of

CLD with respect to time. These rates are calculated and are presented below:

Rate -5.78E-06 -2.74E-05 -4.36E-05

Likewise, rates of change in CLDs are calculated at all the three concentrations. They are tabulated as follows:

Concentration (ppm) 22°C 45°C 70°C 1 -8.92E-07 -1.01E-06 -1. 17E-05 30 -5.16E-06 -2.49E-05 -3.82E-05 60 -5.78E-06 -2.74E-05 -4.36E-05

IV. Regression and Rate Equation:

Using non-linear regression, the data is fit into a rate equation with the help of

'Polymath' software. For the data above, the following rate equation was fit with highest r2 values.

58 A sample polymath report is shown below:

P.QJ;:YN.ATH Rei»Q'" NonlinMr,fiegh~~I

Model: r22 = (k1 *k2*C)/((k1 *C) + k2)

'~~i;bI~li~iti~ig~~;~s IV~lue 95% confidence ..~ ... -..... ~~ .. ~.. "- ...... ~ .. ~ ····-···i·~·······-···-~······~+·~·~~--····~·-···· .. .. k1 ! 1.0E-06 ! 9.S06E-07 1.4S6E-06 ~""~ ... ~.,~-.. ~.... ~-.... ~ ..... ~. ·r··.. ·····•.. ·~ k2 :0.0001 16.3

Nonlinear regression settings Max # iterations = 64

Precision 0.9993547

General f~1Ojl'~i~·~·······

Source data pOints and calculated data points r:r,r';; ,;;;;r ·····················'L~;;·:¢;,·;; tIC Vjr22~;;>;;,;,;A·ir12 caIG,!•• bi~r22 : l1il·0.0000~cis;191'8~494E-~~~~5E~8·i ~ ...... }uu,., .. uu; .",.m« .. ««m ...... _. <...... " .. ••••••• N 0 ••••• +" ~.'N"'.' ... "U"~ .... u'u <.» •••• uuuu ~.; ,2'13010.000005155 15.221E-06 .635E-Osl *0 :O~OOO'0057S4 !S.73E-06 . GE-OS'! .:•.•• }.•.•••..•• j~ ....~ •••~ •••_ ••••••••. ~./ .•~ •• ~ ~ ...... _ .••• ~M •• ~._ ..~I

The above report was generated for the 22°C reaction rates for EPDM - P. The same regression is carried out for 45°C and 70°C as well. Therefore, a rate equation and the respective rate constants are determined in this step.

59 v. Arrhenius equation and activation energy

Arrhenius rate equation,

is used to determine the activation energies of the processes. Natural logarithm of the rate constant vs. the inverse of absolute temperature yields a straight line for which the slope is equal to EalRT. The calculations are tabulated below:

Temperature kl k2 In kl In k2 T(K) Iff 22°C 9.81E-07 6.35E-06 -13.835 -11.967 295 0.00339 45°C 2.68E-06 3.41E-05 -12.83 -10.287 318 0.00314 70°C 1.53E-05 4.40E-05 -11.09 -10.032 343 0.00292

Slope of In kJ vs. 1 / T = -5766.58

Ea = -5766.58 x R = -5766.58 x 0.008314 = 47.94 KJ / Mol

60 CURRICULAM VITAE

NAME . Jahnavi Valleru

ADDRESS 2200 James Pirtle Ct, Apt #1

Louisville, KY 40217

DOB Eluru, India - June 10, 1982

EDUCATION B.Tech, Chemical Engineering

S.V Univesity College of Engineering, India

1999-2003

M.S, Chemical Engineering

University of Louisville, Louisville, KY

2003-2006

61